Re: The problem with Prescott appears to be not only that it is an underwhelming cpu right now, but also that it is very likely to stay such in the foreseeable future. If you disagree with this statement, please, explain how it could be made an "overwhelming" cpu.
Your point is well taken. On the other hand, the gap in performance between Prescott and Athlon 64 is not that huge. In Anand's recent review of the Athlon 64 3800+ vs the Pentium 4 560 (3.6GHz), Intel's chip still performed better in 3D rendering and DIVX encoding, two poster children for needing high performance CPUs. In the regular office tests, the Pentium 4 was ~12% behind in both content creation and business tasks, and about 20% behind in the Quake III compile time.
These are by far not Intel's best, but neither do they represent an unrecoverable gap. I think if Intel added 2M of cache to Prescott and used the 1066MT/s FSB, these parts would easily close the gap. There will be an extreme edition of this variety at 3.73GHz, which will probably give the FX-53 a run for its money, let alone the 3800+ mainstream part.
Let's be honest, Intel is behind in performance right now, but only by margins. They also win some tests in spite of AMD getting good frequency headroom with Athlon 64. The point that I think many will mention is that AMD will scale much higher at 90nm, but I think those of us who know better will admit that they won't have a huge advantage right away. Optimizing a process takes time. AMD might get to 2.6GHz this year, but it will probably take them most of 2005 to get to 3.0GHz. This is normal and expected. Meanwhile, Intel isn't at the end of 90nm, they will eventually hit 4.0GHz when they can tune the process to do so at current thermal envelopes. This isn't trivial, of course, but they haven't had Prescott out a year, yet, so there is still time. You can't claim that a 2.6/2.8GHz Athlon 64 mid-way through next year will be that much farther ahead of a 4.0/4.2GHz Pentium 4/2M/1066MT/s processor than they are today. Therefore, you can't say that Intel has no response for the foreseeable future.
Intel's response is easy: continue to ramp Prescott and derivatives until they can ramp 65nm by the end of 2005, or by Q1 2006 at the latest. They should have enough speed revs and cache sizes left before they get left hopelessly behind in performance. It's the micro-architecture after Prescott that I'm more interested in. I think some time in 2006, Intel will have a lower power micro-architecture that can scale in frequency without being power limited. Maybe this will be based on the Pentium M micro-architecture, or maybe it will be something completely new. That's why I think long term prospects are moot, simply because there isn't enough information, and short term prospects are mediocre. AMD has a good chance to improve GMs with premium parts, but I don't think they are in a position to attack Intel's market share, at least not until they have more manufacturing on hand. Meanwhile, the market is doing well, so both companies will benefit.
That's how I justify an underwhelming CPU as still capable of being successful from the standpoint of Intel.
News 
Market Data 
Discover 
